Announcing Spatial AI: Supercharge your apps with the new spatial intelligence in ADB-S

In this Autonomous Database Learning Lounge session, the Oracle Spatial PM and Autonomous Database PM teams explained how to use the new Spatial AI capabilities in Autonomous Database to improve ML model quality and prediction accuracy.
Register for our bi-weekly sessions and watch recordings at https://bit.ly/adbll-main-y
Slides for this session at: https://bit.ly/adbll-slides-spatial-ai
=== Links from the Session ===
Documentation for Spatial AI
https://docs.oracle.com/en/cloud/paas...
API reference for Spatial AI
https://docs.oracle.com/en/cloud/paas...
=== Links about Autonomous Database ===
Our new Get Started Page is the right place to start: https://bit.ly/adb-get-started
Join our Group on LinkedIn: https://bit.ly/adb-linkedin-grp
Follow us on "X":   / autonomousdw  
Ask questions in:
- Stack Overflow: https://bit.ly/adb-stackoverflow
- Join our External Slack channel and search for 'oracle-autonomous-database': https://bit.ly/odevrel_slack
Outline
David Lapp, Senior Principal Product Manager explained how Spatial AI is a new feature of Oracle Machine Learning for Python (OML4Py) on Autonomous Database Serverless (ADB-S), providing specialized machine learning algorithms that incorporate the effects of location.
Spatial AI can improve ML model quality and prediction accuracy, and an example of spatial regression algorithms that are able to enhance home value predictions by incorporating the influence of neighboring home values was shown.
Spatial algorithms also allow you to detect location patterns, like spatial clustering of traffic accidents. As part of OML4Py on Autonomous Database Serverless, Spatial AI provides a single environment for spatial ML workflows that minimizes data movement to external ML engines, simplifies your architecture, and accelerates time to value.
Video highlights:
01:26 Agenda
02:29 Important links on Autonomous Database
03:06 OML4Py Spatial AI - Location-based predictions and pattern detection
03:58 Key Technologies - OML4Py and ADB-S
05:10 OML4Py Spatial AI - derive insights from spatial data
05:53 Spatial AI Live Demo in Oracle Machine Learning on Autonomous Database
06:48 Identifying the Spatial AI Example Templates available inside Oracle Machine Learning
07:43 OML4Py Spatial AI Run-me-first example
08:37 OML4Py Spatial AI Hotspot Clustering example
11:31 Oracle Spatial - native spatial type and spatial analysis
13:41 OML4Py Spatial AI - Spatial ML algorithms
14:39 Common scenarios addressed by Spatial ML algorithms
20:12 OML4Py Spatial AI - Built into OML4Py on Autonomous Database Serverless
23:25 OML4Py Spatial AI - Features
29:10 OML4Py Spatial AI - Algorithms
31:14 Q&A - Do you have any sample code for spatiotemporal analysis?
32:48 Q&A - Are these algorithms known in the industry or unique to ADB?
35:36 Q&A - How about vessel trip analysis & forecast with AI model
36:37 OML4Py Spatial AI - SpatialDataFrame
42:03 Examples of Spatial AI in use with Los Angeles demographic data summarized by Census Block Group
43:32 SpatialDataFrame proxy to Oracle Spatial Tables
44:19 Spatial operation performed in database
45:43 Feature engineering to add spatial metric
46:59 Fill data gaps with spatial ML algorithm
48:15 Exploratory Analysis and Spatial Autocorrelation
48:43 Detect density clusters
49:20 Detect hotspots and coldspots in data from Oracle Spatial
50:17 SpatialPipeline: scikit-learn Pipeline with support for spatial estimators
50:29 Auto-selection of spatial regression algorithm
51:27 OML4Py Spatial AI - Key Takeaways
52:19 Spatial AI Live demos
57:19 Spatial AI Resources
58:06 Important links on Autonomous Database
58:31 Final thoughts